| CPC B60W 40/02 (2013.01) [B60W 60/001 (2020.02); B60W 2420/403 (2013.01); B60W 2420/408 (2024.01); B60W 2555/20 (2020.02)] | 20 Claims |
|
1. A computer-implemented system, comprising:
one or more non-transitory computer-readable media storing instructions, which when executed by one or more processing units, cause the one or more processing units to perform operations comprising:
estimating locations in a region of a map where features associated with water, snow or ice have a likelihood of being formed under certain adverse weather conditions, the likelihood being greater than a predetermined threshold;
simulating one of the adverse weather conditions;
generating the features associated with the simulated one of the adverse weather conditions at the locations; and
determining a response of a perception stack of an autonomous vehicle (AV) to the adverse weather conditions observed at the locations by testing algorithms, machine learning models, and neural networks of the AV, including the perception stack of the AV, encountering the one of the adverse weather conditions that is simulated.
|
|
9. A method, comprising:
selecting, by a computer-implemented system, a map comprising a region;
identifying, by the computer-implemented system, a topography of the region;
estimating, by the computer-implemented system, a location of at least one of a puddle or a water channel based on the topography;
simulating, by the computer-implemented system, a rainfall condition generating at least one of the puddle or the water channel;
selecting, by the computer-implemented system, a configuration of an autonomous vehicle (AV); and
simulating, by the computer-implemented system, a reaction of the AV according to the configuration to the at least one of the puddle or the water channel in the rainfall condition by running algorithms, machine learning models, and neural networks of the AV, including a perception stack of the AV, encountering the at least one of the puddle or the water channel in the rainfall condition that is simulated.
|
|
14. A method, comprising:
selecting, by a computer-implemented system, a map comprising a region;
identifying, by the computer-implemented system, a topography of the region;
simulating, by the computer-implemented system, a snowfall condition generating snow accumulation;
estimating, by the computer-implemented system, a location of snow accumulation based on the topography and the snowfall condition;
selecting, by the computer-implemented system, a configuration of an AV; and
simulating, by the computer-implemented system, a reaction of the AV according to the configuration to the snow accumulation in the snowfall condition by running algorithms, machine learning models, and neural networks of the AV, including a perception stack of the AV, encountering the at least one of the puddle or the water channel in the rainfall condition that is simulated.
|